Skip to main content
Loading navigation
Back to challenges

Multi-Omics ALS Biomarker Longitudinal Fusion Pipeline with Pydantic AI

NIH news highlights blood-based biomarker breakthroughs for predicting pre-symptomatic ALS symptom emergence. In this challenge, participants will build a multi-omics longitudinal data integration pipeline using Pydantic AI. The agent will ingest serum proteomics data (NfL, pNfH), transcriptomic panels, and clinical histories to predict disease conversion timelines with strict schema-enforced output constraints.

Data ScienceHosted by Vera
Sign in free to start

What you are building

Build a type-safe multi-omics ALS biomarker longitudinal prediction pipeline using Pydantic AI.

How work is evaluated

Evaluation

Evaluates hazard ratio estimation precision and time-to-conversion prediction.

Datasets

Public datasets, plus the private datasets you have access to.

How submissions are scored

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

check_valid_hazard_ratio

Ensure hazard ratio is positive float

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

hazard_ratio_mae

Mean Absolute Error of estimated hazard ratio against longitudinal survival ground truth • target: 0.25 • range: 0-5

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

  • Construct Pydantic AI agents for multi-modal omics longitudinal integration

  • Ingest plasma proteomics (NfL) and RNA expression panels

  • Apply time-to-event survival models (Cox PH, Random Survival Forests) via Python dependencies

  • Enforce strictly validated patient conversion prognosis schemas

Resources

Dataset notes

Synthetic multi-omics panels (plasma NfL, pNfH, miRNA expression) paired with conversion timeline outcomes.

Challenge setup

Evaluates hazard ratio estimation precision and time-to-conversion prediction.

Challenge input

JSON containing nfl_pg_ml, pnfh_pg_ml, mirna_133b_expr

Pydantic AI

Mandatory Pydantic AI framework

Evaluated output

JSON containing hazard_ratio (float), predicted_conversion_window_months (string)

Evaluation checks
  • Ensure hazard ratio is positive float
  • Mean Absolute Error of estimated hazard ratio against longitudinal survival ground truth • target: 0.25 • r...
Configured evaluation evidence
  • Benchmark: OmicsSurvival-Bench
  • Hazard Ratio Mae target: 0.25
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Shows the first task, up to three tools, and up to three evidence items.

View technical recipe

Configured tools

Action Space
  • Pydantic AI · Required
  • Merge · Optional
  • Llama 3.2 · Optional
Orchestration
  • Pydantic AI · Required

Evaluation contract

  • check_valid_hazard_ratio · Weight 1
  • hazard_ratio_mae · Weight 1

Metrics

  • hazard_ratio_mae · Target 0.25

Benchmarks

  • OmicsSurvival-Bench
  • biotech

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing nfl_pg_ml, pnfh_pg_ml, mirna_133b_expr

    Output: JSON containing hazard_ratio (float), predicted_conversion_window_months (string)

    Execution harness: Python

Evaluation tools

  • Pydantic AI · Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Multi-Omics ALS Biomarker Longitudinal Fusion Pipeline with Pydantic AI